ABSTRACT: Extraction of coconut oil with a pure culture ofLactobacillus plantarum 1041 IAM was investigated. Grated co-conut meat and water at 30, 50, and 70°C were mixed in vari-ous ratios (1:1, 1:2, and 1:3) and allowed to settle for 2–6 h. Themost efficient coconut cream separation was obtained at the 1:1ratio of grated coconut meat to water at 70°C, followed by 6 hsettling time. Fermentation was then conducted on coconutcream emulsion with the sample from 1:1 ratio, 70°C, and 6-hsettling time. Oil yield from the fermentation process with 5%inoculum of L. plantarum 1041 IAM after 10 h at 40°C was95.06% Quality characteristics of the extracted oil were as fol-lows: moisture content, 0.04%; peroxide value, 5.8 meq oxy-gen/kg; anisidine value, 2.10; free fatty acid, 2.45%; iodinevalue, 4.9; and color, 0.6 (Y + 5R). Extraction of coconut oilfrom coconut meat with L. plantarum 1041 IAM was signifi-cantly improved in both oil yield and quality over the traditionalwet process.JAOCS 74, 1115–1119 (1997).

KEY WORDS: Coconut cream, coconut meat, fermentation,Lactobacillus plantarum 1041 IAM, oil yield, quality character-istics.

About 10% of the total oils and fats entering the world mar-ket is coconut oil (1). Several methods are currently practicedfor removing oil from either fresh coconut meat or copra(dried coconut kernel). These technologies include the (i) wetprocess, (ii) the dry process, and (iii) solvent extraction. (i)The wet process can be carried out by grinding coconut meatwith water and filtering it to produce coconut milk or coconutcream. This emulsion contains protein and coconut oil, whichcan be separated either through common kitchen utensils orhydraulic presses. However, coconut oil extraction by thesewet process techniques has not been commercially successful(2–4). (ii) The dry process is the present commercial tech-nique for coconut oil extraction. Copra is cleaned, ground,steamed, and pressed through an expeller for coconut oil ex-traction. This extracted oil is further purified by neutraliza-tion, bleaching, and deodorization to remove free fatty acids(FFA), odors, flavors, and pigments. (iii) Solvent extraction ispossible with an appropriate solvent, such as benzene or

n-hexane. Even though oil recovery is high, the process israrely applied owing to its high risk and high investment cost.

Usually, the recovery of coconut oil by the traditional wetprocess is low, about 30–40% (5). Moreover, the oil obtainedis of poor quality owing to the high moisture content (MC)dark color, and short shelf life (6). The process is also energy-and time-consuming. On the other hand, the traditionalmethod is easy to handle, and the extracted coconut oil has apleasant aroma and low FFA (7). Several workers have inves-tigated alternative wet extraction methods to recover coconutoil. Che Man et al. (8) studied the use of 0.1–0.4% acetic acid(25%) for coconut oil extraction and showed a recovery of58.3–60.3% of good-quality oil. In another study, Che Man etal. (9) obtained a yield of 73.8% of good-quality oil with anenzyme mixture at 1% (w/w) each of cellulase, α-amylase,polygalacturonase, and protease at pH 7.0 and an extractiontemperature of 60°C. The extracted oils in these studies re-quired no further purification to meet the quality of the pro-posed International Standard by the Asian and Pacific Co-conut Community (4). An earlier study by McGlone et al.(10), in which a mixture of polygalacturonase, α-amylase,and protease were used on diluted coconut paste, obtained an80% yield of good-quality oil, compared to the Official Mex-ican Standards (10). Suhardiyono (11) investigated the use ofbaker’s yeast to extract coconut oil. The mixed culture of bak-er’s yeast grew in coconut milk and broke the emulsion intogood-quality oil. Moreover, the yield of extracted oil wasdouble that of the traditional wet process. Based on this work,we postulated that higher recovery of coconut oil could be ob-tained by using pure cultures of microorganisms. Therefore,the objectives of this study were to investigate the effect of apure culture of Lactobacillus plantarum 1041 IAM to breakthe coconut cream emulsion for separating the oil and to de-termine the yield and quality characteristics of the extractedoil.

MATERIALS AND METHODS

Materials. Fresh coconut of the Mawar variety was obtainedfrom Universiti Pertanian Malaysia Farm (Selangor,Malaysia). A pure culture of L. plantarum strain 1041 IAMwas obtained from the University of Tokyo, Japan. Commer-cial coconut oil was purchased from Kilang Minyak Tanjung

Copyright © 1997 by AOCS Press 1115 JAOCS, Vol. 74, no. 9 (1997)

*To whom correspondence should be addressed.E-mail: yaakub@fsb.upm.edu.my.

Extraction of Coconut Oilwith Lactobacillus plantarum 1041 IAM

Y.B. Che Mana,*, M.I.B. Abdul Karimb, and C.T. Tenga

Departments of aFood Technology and bBiotechnology, Universiti Pertanian Malaysia, 43400 UPM, Serdang, Malaysia

Karang Sdn. Bhd. (Selangor, Malaysia). All chemicalreagents were of analytical grade and obtained from BDHChemical Ltd. (Poole, England).

Extraction of coconut cream. The extraction of coconutmilk was carried out as follows: Grated coconut meat andwater at 30°C were mixed in proportions of 1:1, 1:2, and 1:3.The mixture was kneaded manually for 5 min, and the milkwas extracted, squeezed, and strained through a layer ofcheesecloth. The coconut milk obtained was then left to settlefor 2, 4, and 6 h. The samples based on the ratio of 1:1 wereused to determine the effect of different settling times (2, 4,and 6 h) and temperatures (30, 50, and 70°C) on the oil ex-traction yield. Coconut milk was then allowed to settle andseparate into two layers: the upper cream emulsion layer,which was thick and dense, and the lower aqueous layer,which contained mainly water and was drained off.

Treatment of coconut cream by chemicals and heat. Thefreshly extracted coconut cream was treated chemically as de-scribed by Bhowmik and Marth (14) with some modifica-tions. The coconut cream (200 mL), containing 1 g of 30%hydrogen peroxide (H2O2), was incubated at 40°C for 2 h.Catalase (10.5 mg) was then added to decompose the H2O2.Coconut cream was further incubated at 30°C for 3 h. A pre-liminary experiment was carried out to determine the degreeof stability of the product by a plate count. As a result of theperoxide/catalase treatment, the bacterial plate count was re-duced to 3870 cells/mL coconut cream compared to 4.01 ×108 cells/mL coconut cream in the original sample.

Fermentation. A pure culture of L. plantarum strain 1041IAM was transferred to the MRS medium for cell activationat 30°C for 48 h (13). Pellets of L. plantarum were obtainedafter centrifugation with phosphate buffer saline (PBS) solu-tion several times. Dilution was carried out in PBS with dif-ferent concentrations between 10−1 and 10−6. These dilutionswere later checked with a spectrophotometer at 540 nm to ob-tain the appropriate optical density (OD). The number of lac-tic acid bacteria cells was determined by hemacytometer atthe selected and specified OD (0.01–0.06). Coconut creamwas divided into two parts. Portion of the cream were trans-ferred to fermentation vessels where 1, 3, and 5% inocula ofL. plantarum 1041 were added; another portion was used as acontrol. Fermentation was carried out from 2 to 10 h at 40°C(12).

Determination of lactic acid. Coconut cream (10 mL) wasaugmented with five drops of 0.5% phenolphthalein, followedby titration with 0.1 N NaOH solution. The acid produced byL. plantarum 1041 IAM was calculated as lactic acid (%,wt/vol) with the formula: 1 mL 0.1 N NaOH = 0.009 g lacticacid.

Oil recovery. The oil recovery was calculated based on theinitial oil content of the coconut meat as determined by theSoxhlet method of AOAC (15) and the direct weight mea-surement of oil obtained after extraction.

Analyses of oil quality. MC and percentage (%) FFA weremeasured according to AOAC methods (15). Peroxide (PV)and iodine (IV) values were measured according to British

Standard No. 684 (16). The anisidine value (AnV) and fattyacid composition (FAC) were measured according to PORIMTest Methods (17). Color was measured by Lovibond Tin-tometer (Model E) according to British Standard No. 684(16).

Statistical analysis. The data were analyzed by analysis ofvariance techniques. Means that were significantly differentat a 5% level of probability (P < 0.05) were further separatedby Duncan’s Multiple Range Test (18).

RESULTS AND DISCUSSION

Effect of coconut meat/water ratio on oil yield. The effect ofdifferent amounts of water at 30°C added to coconut meat onextraction yield of oil is shown in Table 1. The results showthat the 1:1 ratio gave the highest oil yield, compared to 1:2and 1:3. As the ratio of water added was increased, the oilcontent decreased. Higher proportions of water increased thedilution effect and therefore decreased the oil yield in coconutcream from 37.09 to 32.50 and 28.48% in 1:1, 1:2, and 1:3,respectively, after 2 h of settling time. The same trend wasfound for 4- and 6-h settling times. Banzon et al. (1) foundthat the composition of coconut cream is largely based on theamount of water added for the extraction of oil. This resultshowed that adding less water contributed to a higher propor-tion of oil and is in agreement with the finding of Banzon andco-workers.

Effect of water temperature and settling times on oil yield.Table 2 shows that as water temperature was increased from30 to 70°C, there was a significantly increased (P < 0.05) oilyield for the 1:1 ratio of coconut meat/water when followedby 2–6 h settling times. The oil recovery from coconut creamwith 70°C water was 46.23%, while those of water at 30 and50°C were 37.09 and 40.83%, respectively. The most effec-tive time of settling the coconut milk was 6 h, which yielded51.33% oil when the water was at 30°C. The yield was fur-ther increased to 83.88% with water at 70°C. Therefore, along settling time is required at higher temperature to effi-ciently separate the cream emulsion so that higher oil yield isobtained.

Oil yield after fermentation. Based on the previous experi-ments, coconut cream from the 1:1 coconut meat/water ratio(wt/vol) was used for fermentation study. The concentratedcoconut cream contained lower proportions of skim milk andtherefore reduced the volume of inoculum added. Puertollanoet al. (12) also reported that the optimum dilution for rapid

1116 Y.B. CHE MAN ET AL.

JAOCS, Vol. 74, no. 9 (1997)

TABLE 1Effect of Coconut Meat/Water Ratio on Oil Yield (%) at 30°Ca

Settling time (h)Coconut meat/water 2 4 6

1:1 37.09a 43.04a 51.33a

1:2 32.50b 38.38a,b 41.70b

1:3 28.48c 36.18b 41.61b

aMean of three readings. Means in a column followed by different super-script letters (a–c) are different (P < 0.05).

de-emulsification and separation of oil was coconut milk of1:1 coconut meal/water ratio (wt/vol). Lower proportions ofwater, for example 1:0, can hinder the extraction of coconutoil, protein, and other extractable materials owing to insuffi-cient fluidity for grinding. However, a higher proportion ofwater increased the time required to break the emulsion, suchas a ratio of 1:3 for sample and volume of inoculum.

Coconut cream contained a high bacterial count, and thepredominant organisms were gram-positive bacteria withlong and short rods (12). Therefore, H2O2 was added to thecream, followed by heating at 40°C to kill the backgroundmesophilic bacteria, to reduce competition between the back-ground bacteria and L. plantarum strain 1041 IAM. Fermen-tation in an Erlenmeyer flask provided a microaerophilic con-dition for growth of the lactobacilli.

The formation of lactic acid during fermentation is pre-sented in Tables 3 and 4. Inoculation of coconut cream withL. plantarum strain 1041 IAM resulted in rapid breaking ofthe emulsion. We believe that L. plantarum used glucose, theonly sugar present in coconut cream, at 1.23% (14), forgrowth and consequently caused the production of lactic acid.Lactic acid increased as the amount of L. plantarum inocu-lated increased to 5%, indicating that a higher rate of fermen-tation occurred in samples with a ratio of 1:1 coconutmeat/water at 30°C and 2-h settling time. The lactic acid pro-duced ranged from 0.207 to 0.368% in inoculated coconutcream, compared to lower lactic acid levels, 0.18 to 0.315%,without inoculation. However, the lactic acid produced in thetreatment of 1:1 coconut meat/water with a water temperatureof 70°C, followed by 6 h settling time, ranged from 0.0351 to0.0522%, which was lower than the former treatment. Precip-itation of soluble protein in the interfacial film occurred as aresult of lactic acid formation during fermentation. Lacticacid destabilized the protein and caused water to be released(11).

Oil recovery. The effect of various concentrations of L.plantarum strain 1041 IAM during fermentation at 40°C and10 h incubation time on oil yield is presented in Table 5.Shorter incubation times were less effective. The amount ofoil obtained ranged from 37.09 (control) to 59.91% (after fer-mentation with water at 30°C and 2h settling time) and 83.88(control) to 95.06% (after fermentation with water at 70°Cand 6h settling time). The highest oil yield was 95.06% at 5%inoculum with a sample of 1:1 ratio, 70°C and 6 h settling

time. This study showed that oil recovery was improved afterfermentation, compared to the traditional method, which wasabout 30–40%. However, at 70°C and 6 h settling time, oil re-covery was tremendously improved.

Quality characteristics of extracted coconut oil. The FACof the extracted coconut oil is presented in Table 6. The re-sults are in agreement with the Codex Alimentarius Commis-sion International Standard (CACIS) value (19). As a com-parison, the FAC of commercial coconut oil was in closeagreement with the extracted oil. However, extracted coconutoil showed higher total saturated fat, or 92.43%, comparedwith 88.45% in commercial coconut oil. The high degree ofsaturation caused a higher resistance to rancidity (20).

The color of the oil was unchanged at 0.3 (Y + 5R), asshown in Table 7, for low-temperature and 2-h settling timetreatment and 0.6 (Y + 5R) for high-temperature and 6-h set-tling time treatment, as shown in Table 8. PV in the extractedoil ranged from 2.55 to 5.15 meq oxygen/kg, compared to8.95 meq oxygen/kg in commercial oil. A 5% inoculum con-tributed to the lowest PV, 2.55 meq oxygen/kg, compared to3.50 and 4.05 meq oxygen/kg in 3 and 1% inoculum in low-temperature short settling time. For the high-temperature, 6-hsettling treatment, a 5% inoculum gave a PV of 5.8, comparedto 7.2 and 7.6 meq oxygen/kg in 3 and 1% inoculum, which

EXTRACTION OF COCONUT OIL 1117

JAOCS, Vol. 74, no. 9 (1997)

TABLE 2Effect of Settling Times and Water Temperatures Based on 1:1 Ratioof Coconut Meat/Water on Oil Yield (%)a

Settling time Water temperature (°C)(h) 30 50 70

2 37.09c,f 40.83c,e 46.23c,d

4 43.04b,f 61.42b,e 70.63b,d

6 51.33a,f 75.76a,e 83.88a,d

aMeans of three readings. Means in a column followed by different super-script letters (a–c) are different (P < 0.05). Means in a row followed by dif-ferent superscript (d–f) letters are different (P < 0.05).

TABLE 3Effect of Incubation Time and Inoculum Level on Formation of LacticAcid (%)a During Fermentation of Coconut Milk Extracted at 30°C,Followed by Settling (2 h)b

Inoculum Incubation time (h)(%) 2 4 6 8 10

0 0.180d,h 0.190d,h 0.20c,g 0.225b,g 0.315a,f

1 0.20d,g 0.288c,g 0.315b,f 0.322b,f 0.330b,g

3 0.225b,f 0.304b,f 0.336a,e 0.342c,f 0.352c,g

5 0.243d,e 0.323a,e 0.350a,e 0.358c,e 0.368b,e

aMeans of three readings. Means in a row followed by different superscriptletters (a–d) are different (P < 0.05). Means in a column followed by differ-ent superscript letters (e–h) are different (P < 0.05). bFermentation with Lac-tobacillus plantarum was carried out at 40°C. Coconut milk was derived bya water extraction of grated coconut meat (1:1, coconut meat/water).

TABLE 4Effect of Incubation Time and Inoculum Level on Formationof Lactic Acid (%)a During Fermentation of Coconut MilkExtracted at 70°C, Followed by Settling (6 h)b

Inoculum Incubation time (h)(%) 2 4 6 8 10

0 0.0342e,i 0.0378d,i 0.0405c,h 0.045b,i 0.0477a,i

1 0.0351e,h 0.0387d,h 0.0423c,i 0.0477b,h 0.0486a,h

3 0.0369e,g 0.0405d,g 0.0432c,g 0.0495b,g 0.0504a,b

5 0.0459e,f 0.0477d,f 0.0495c,f 0.0513b,f 0.0522a,f

aMean of three readings. Means in a row followed by different superscriptletters (a–e) are different (P < 0.05). Means in a column followed by differ-ent superscript letters (f–i) are different (P < 0.05).bFermentation with Lactobacillus plantarum was carried out at 40°C. Co-conut milk was derived by a water extraction of grated coconut meat (1:1,coconut meat/water).

was higher than the former treatment. However, the maxi-mum PV according to CACIS is 10 meq oxygen/kg. Althoughthe oil quality of the latter treatment was not so good as theformer, it was still within the range for standard good oil. Per-oxides are the primary products formed by oxidation of theoil. They are unstable and break down to many types of sec-ondary products. The deterioration of oils cannot be measuredaccurately by using this method alone (21) because a decreasein PV does not necessarily indicate the oil is in good condi-

tion. Theoretically, coconut oil should show a low rate of oxi-dation because it contains low levels of unsaturated fattyacids.

The comparison of AnV between extracted coconut oil andcommercial oil is presented in Tables 7 and 8. There was asignificant difference (P < 0.05) in AnV as the percentage in-oculum was increased. Commercial oil had higher AnV, 0.73compared to extracted coconut oil, which ranged from 0.12to 0.163 in low-temperature short settling time and from 2.10to 4.59 in high-temperature long settling time. According toRussell (22), as a rule of thumb for good-quality oils, the AnVshould be less than 10. Inoculum at 5% showed the lowestAnV, 0.12 and 2.10, for low temperature, 2-h settling time andhigh-temperature, 6-h settling time, respectively.

Tables 7 and 8 also show the results obtained for FFA andMC. The higher the MC in the oil, the higher the percentageof FFA. MC in extracted coconut oil after fermentationranged from 0.013 to 0.018% in low-temperature short set-tling time treatment, which was lower than the 0.034% incommercial oil. Inoculum at 5% showed the lowest MC at0.013%, compared to 3 and 5% inoculum, whereas uninocu-lated coconut cream produced oil with 0.023% MC, which ishigher than the inoculated coconut cream. MC in high-tem-perature longer settling time treatment ranged from 0.039 to0.057%. Inoculum at 5% showed the lowest MC, 0.039%,compared to 0.044 and 0.049% in 3 and 1% inoculum, re-spectively.

FFA content ranged from 0.035 to 0.05% in low-tempera-ture short settling time treatment, 0.07% in uninoculated co-conut cream, and 1.55% in commercial oil, indicating that theextracted oil has a better quality. However, high-temperaturelonger settling time treatment showed a higher percentage ofFFA, 2.45, 2.76 and 3.04% at 5, 3, and 1% inoculum, respec-tively, compared to low-temperature short time treatment.Maximum percentage FFA, according to CACIS, is 4, whichmeans that an oil with a higher value is considered rancid.Production of FFA through hydrolysis is related to the sec-ondary products of oxidation (23).

1118 Y.B. CHE MAN ET AL.

JAOCS, Vol. 74, no. 9 (1997)

TABLE 5Effect of Inoculum Level, Settling Time, and Extraction Temperatureon Oil Yield (%)a Following Incubationb

Inoculum (%) 1:1 ratio, 30°C, 2 h settling 1:1 ratio, 70°C, 6 h settling

0 37.09c 83.88c

1 54.49b 90.46b

3 57.32a,b 93.64a,b

5 59.91a 95.06a

aMean of three readings. Means in a row followed by different superscriptletters (a–c) are different (P < 0.05).bFermentation with Lactobacillus plantarum IAM 1041 was carried out at40°C on coconut milk derived by water extraction of grated coconut meat(1:1, coconut meat/water).

TABLE 6Fatty Acid Composition (%) of Standard, Extracted,and Commercial Coconut Oil

Fatty acid Standard oila Extracted oil Commercial oil

Caproic (C6) <1.2 n.d. n.d.Caprylic (C8) 3.4–15 6.32 5.78Capric (C10:0) 3.2–15 5.59 5.39Lauric (C12:0) 41–46 49.69 49.49Myristic (C14:0) 13–23 19.94 18.33Palmitic (C16:0) 4.2–12 8.83 10.38Stearic (C18:0) 1–4.7 2.06 2.08Oleic (C18:1) 3.4–12 6.36 9.01Linoleic (C18:2) 0.9–3.7 1.206 2.545aReference 19; n.d., not detected.

TABLE 7Effect of Inoculum Level of Lactobacillus plantarum 1041 IAMon Quality Characteristics of Oil Extracted with 30°C Water,Followed by Settling (2 h)a,b,c

Inoculum(%) Color PV AnV FFA MC IV

0 0.3a 5.15b 0.49b 0.075b 0.023b 5.89b

1 0.3a 4.05c 0.163c 0.05c 0.018c 3.80c

3 0.3a 3.50c 0.140d 0.04d 0.016d 3.70d

5 0.3a 2.55e 0.120e 0.035e 0.013e 3.10e

Cd 0.3a 8.95a 0.73a 1.55a 0.034a 8.71a

aReported results are means of three readings. Means in a column followedby different superscript letters (a–e) are different (P < 0.05).bCoconut meat was extracted with water (1:1, coconut meat/water) at 30°Cas described in the Materials and Methods section, then allowed to settlefor 2 h. Samples were then inoculated with L. plantarum and incubated for10 h at 40°C.cColor, determined as Y + 5R; PV, peroxide value (meq oxygen/kg); AnV,anisidine value (1,000 × abs); FFA, free fatty acid content (%); MC, mois-ture content (%); IV, iodine value (g I/100 g sample). dC: Commercial oil.

TABLE 8Effect of Inoculum Level of Lactobacillus plantarum 1041 IAMon Quality Characteristics of Oil Extracted with 70°C Water,Followed by Settling (6 h)a,b,c

Inoculum(%) Color PV AnV FFA MC IV

0 0.6a 9.6a 4.59a 3.48a 0.057a 7.61b

1 0.6a 7.6c 2.56b 3.04b 0.049b 7.61b

3 0.6a 7.2d 2.21c 2.76c 0.044c 6.73c

5 0.6a 5.8e 2.10d 2.45c 0.039d 4.95d

Cd 0.3a 8.95a 0.73a 1.55a 0.034a 8.71a

aReported results are means of three readings. Means in a column followedby different superscript letters (a–e) are different (P < 0.05).bCoconut meat was extracted with water (1:1, coconut meat/water) at 70°Cas described in the Materials and Methods section, then allowed to settlefor 6 h. Samples were then inoculated with L. plantarum and incubated for10 h at 40°C.c,dFor abbreviations see Table 7.

The IV of extracted coconut oil and commercial oil isshown in Tables 7 and 8. The IV indicates the quality of oilbased on the unsaturation of fatty acids. The results obtainedranged from 3.1 to 3.8 in low-temperature short settling timetreatment and 4.95 to 7.61 in high-temperature long settlingtime treatment. The IV of commercial oil was 8.71, which iswithin the standard range of CACIS of 6–11 (19). A lower IVcan be reflected by lower amounts of unsaturated fatty acidsfrom the FAC profile in Table 6. Therefore, the stability of therecovered oil was better than the uninoculated coconut oil andcommercial oil in terms of its resistance to oxidation.

The technique for coconut oil extraction by fermentationwith a pure culture of Lactobacillus plantarum 1041 IAMshowed a higher oil extraction yield over the previous tech-niques of acetic acid and aqueous enzymatic extraction as re-ported by the senior author and co-workers (8,9). All thesetechniques had shown a significant improvement over the tra-ditional wet process (30–40%) that is currently practiced inmany coconut-producing countries.

ACKNOWLEDGMENT

This is a contribution of Project No. 1-50218-90-01 of UniversitiPertanian Malaysia.

REFERENCES

1. Banzon, J.A., O.M. Gonzales, S.Y. de Leon, and P.C. Sanches,Coconut as Food. Philippines, Coconut Research and Develop-ment, Diliman, Quezon City, Philippines, 1990, pp. 3–140.

2. Child, R., Coconut, 2nd edn., Longman, London, 1974.3. Hagenmaier, R., Aqueous Process, Philippines Coconut Re-

search and Development Foundation Inc., Quezon City, 1980.4. Thampan, P.K., Handbook of Coconut Palms, 2nd edn., Oxford

and IBH Publishing Company Inc. New Delhi-Bombay-Cal-cutta, 1984.

5. Thieme, J.G., Coconut Oil Processing, United Nations Food andAgriculture Organization, Rome, 1968.

6. Hagenmaier, R.D., C.M. Cater, and K.F. Mattil, Aqueous Pro-cessing of Fresh Coconuts for Recovery of Oil and CoconutSkim Milk, J. Food Sci. 38:516–518 (1973).

7. Loo, T.G., Small Scale and Home Processing of Fresh Coconut(Oil Manufacture) and Utilisation of By-Products, Royal Tropi-cal Institute, Amsterdam, 1982.

8. Che Man, Y.B., Suhardiyono, A.B. Asbi, and M.N. Azudin,Acetic Acid Treatment of Coconut Cream in Coconut Oil Ex-traction, ASEAN Food J. 7:38–42 (1992).

9. Che Man, Y.B., Suhardiyono, A.B. Asbi, M.N. Azudin, and L.S.Wei, Aqueous Enzymatic Extraction of Coconut Oil, J. Am. OilChem. Soc. 73:683–686 (1996).

10. McGlone, O.C., A.L.M. Canales, and J.V. Carter, Coconut OilExtraction by a New Enzymatic Process, J. Food Sci.51:695–697 (1986).

11. Suhardiyono, Effects of Extraction Methods on Recovery Qual-ity, Storage Stability and Frying Characteristic of Coconut Oil.M.Sc. Thesis, Faculty of Food Sci. and Biotechnology, Univer-siti Pertanian Malaysia, 1992, pp. 51–68.

12. Puertollano, C.L., J. Banzon, and K.H. Steinkraus, Separationof the Oil and Protein Fractions in Coconut by Fermentation, J.Agric. Food Chem. 18:243–248 (1970).

13. de Man, J.C., M. Rogosa, and M.E. Sharpe, A Medium for theCultivation of Lactobacilli, J. Appl. Bacteriol. 23:130 (1960).

14. Bhowmik, T., and E.H. Marth, Protease and Peptide Activity ofMicrococcus Species, J. Dairy Sci. 71:2358–2365 (1988).

15. Association of Official Analytical Chemists, Official Methodsof Analysis, 14th edn., AOAC, Washington, D.C., 1984.

16. British Standard, Methods of Analysis of Oils and Fats, BritishStandard Institution, London, 1976.

17. PORIM, PORIM Test Methods, Palm Oil Research Institute ofMalaysia, Ministry of Primary Industries, Kuala Lumpur,Malaysia, 1988, 170 pp.

18. Little, L.M., and J.P. Hill, Agricultural Experimentation, JohnWiley & Sons, New York, 1977.

19. Young, F.V.K., Palm Kernel and Coconut Oil: Analytical Char-acteristics Process Technology and Uses, J. Am. Oil Chem. Soc.60:326A–331A (1983).

20. Ropati, F.N., Studies of the Composition of Coconut Products,M.Sc. Thesis, University of Western Sydney, Australia, 1992,pp. 12–15.

21. Wan Hussin, W.R., Comparison of Frying Performance betweenPalm Olein and Coconut Oil, B.Sc. Thesis, Faculty of Food Sci-ence and Biotechnology, Universiti Pertanian Malaysia, 1992,75 pp.

22. Russell, J.B., Measurement of Rancidity, Rancidity in Food, 2ndedn., Elsevier Applied Science Publishing Co., Inc., London,1989.

23. Berger, K.G., The Practice of Frying, PORIM Technology. No.9, Palm Oil Research Institute of Malaysia, Kuala Lumpur,1984.

[Received January 2, 1997; accepted May 3, 1997]

EXTRACTION OF COCONUT OIL 1119

JAOCS, Vol. 74, no. 9 (1997)